Suppression of LPS-induced matrix-metalloproteinase responses in macrophagesexposed to phenytoin and its metabolite,5-(p-hydroxyphenyl-), 5-phenylhydantoinRyan Serra1, Abdel-ghany Al-saidi1, Nikola Angelov2, Salvador Nares1*
Abstract
Background: Phenytoin (PHT) has been reported to induce gingival (gum) overgrowth (GO) in approximately 50%of patients taking this medication. While most studies have focused on the effects of PHT on the fibroblast in thepathophysiology underlying GO, few studies have investigated the potential regulatory role of macrophages inextracellular matrix (ECM) turnover and secretion of proinflammatory mediators. The aim of this study was toevaluate the effects of PHT and its metabolite, 5-(p-hydroxyphenyl-), 5-phenylhydantoin (HPPH) on LPS-elicitedMMP, TIMP, TNF-a and IL-6 levels in macrophages.
Methods: Human primary monocyte-derived macrophages (n = 6 independent donors) were pretreated with 15-50 μg/mL PHT-Na+ or 15-50 μg/mL HPPH for 1 hour. Cells were then challenged with 100 ng/ml purified LPS fromthe periodontal pathogen, Aggregatibacter actinomycetemcomitans. Supernatants were collected after 24 hours andlevels of MMP-1, MMP-2, MMP-3, MMP-9, MMP-12, TIMP-1, TIMP-2, TIMP-3, TIMP-4, TNF-a and IL-6 determined bymultiplex analysis or enzyme-linked immunoadsorbent assay.
Results: A dose-dependent inhibition of MMP-1, MMP-3, MMP-9, TIMP-1 but not MMP-2 was noted in culturesupernatants pretreated with PHT or HPPH prior to LPS challenge. MMP-12, TIMP-2, TIMP-3 and TIMP-2 were notdetected in culture supernatants. High concentrations of PHT but not HPPH, blunted LPS-induced TNF-aproduction although neither significantly affected IL-6 levels.
Conclusion: The ability of macrophages to mediate turnover of ECM via the production of metalloproteinases iscompromised not only by PHT, but its metabolite, HPPH in a dose-dependent fashion. Further, the preferentialdysregulation of macrophage-derived TNF-a but not IL-6 in response to bacterial challenge may provide aninflammatory environment facilitating collagen accumulation without the counteracting production of MMPs.
BackgroundDrug-induced gingival (gum) overgrowth (DIGO) iswidely recognized as a common unwanted sequelaeassociated with a variety of medications. Among these,the antiepileptic agent, PHT (Dilantin®), has beenreported to induce gingival overgrowth (GO) in approxi-mately 50% of patients taking this medication [1,2]. PHTis a hydantoin-derivative anticonvulsant that exerts its
anticonvulsant properties by stabilizing neuronal cellmembranes to the action of sodium, potassium, and cal-cium. The drug also affects the transport of calciumacross cell membranes and decreases the influx of cal-cium ions across membranes by decreasing membranepermeability and blocking intracellular uptake [3]. PHTis primarily metabolized by liver cytochrome P450enzymes, particularly CYP2C9 and CYP2C19 [4] to formenantiomers of 5-(4-hydroxyphenyl-),5-phenylhydantoin(HPPH) which in addition to PHT, have been implicatedin the pathogenesis of DIGO [5,6].* Correspondence: [email protected]
1Department of Periodontology, School of Dentistry, University of NorthCarolina at Chapel Hill, Chapel Hill, North Carolina, USAFull list of author information is available at the end of the article
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
While most studies have focused on the role of thefibroblast [7-10], it is likely that other cells contribute tothe pathogenesis of DIGO. In particular, tissue macro-phages, present in elevated numbers within gingival tis-sues, possibly in response to accumulation of the plaquebiofilm [2,11], may play a role in pathogenesis. Theselong-lived, multifaceted cells, strategically poised alongportals of entry, perform numerous functions of vitalimportance to the host. In addition to their key role inimmunity [12], the macrophage is recognized as themajor mediator of normal connective tissue turnoverand maintenance, as well as for orchestrating repair dur-ing wound healing [13-18]. It has a dualistic role toreceive, amplify, and transmit signals to fibroblasts,endothelial cells, and vascular smooth muscle cells byproducing pro-inflammatory and catabolic cytokines.However, during tissue turnover and wound healing itsecretes anabolic peptide growth factors [12]. Given thisduality of function, any perturbation can lead to patho-logical processes. We have demonstrated that the clini-cal presentation of PHT-induced gingival overgrowth isassociated with a specific macrophage phenotype char-acterized by high expression levels of IL-1b and PDGF-B [11,19] suggesting that this drug-induced macrophagephenotype could contribute to the pathogenesis ofDIGO. These cellular attributes might explain thedichotomy of the lesion where there is both periodontalinflammation typically associated with connective tissuecatabolism paradoxically juxtaposed with gingival over-growth,- a clear anabolic signal of wound repair andregeneration.As tissue homeostasis requires the proper balance of
metabolism and catabolism, it is possible that macro-phage-derived cytokines, MMPs and TIMP levels arealtered in response to PHT and HPPH. Here we investi-gated the effects of these agents on production ofMMPs, TIMPs, and pro-inflammatory cytokines inhuman monocyte-derived macrophages and report thatindeed, PHT and HPPH significantly modulate macro-phage MMP and cytokine protein levels in response topurified LPS from the periodontal pathogen, Aggregati-bacter actinomycetemcomitans.
MethodsMonocyte isolation and macrophage differentiationPeripheral blood mononuclear cells were obtained fromcommercially-available buffy coats (Oklahoma BloodInstitute, Oklahoma City, OK, USA) derived fromhealthy donors by density gradient centrifugation usingFicoll-paque (Amersham, Uppsala, Sweden). Six inde-pendent cultures were obtained from 6 independentdonors. Monocytes were isolated using CD14 MicroBe-ads (Miltenyi Biotec, Auburn, CA, USA) according tomanufacturer’s instructions and cultured as previously
described [12,20,21]. Briefly, isolated monocytes wereplated onto duplicate 12-well tissue culture-treatedplates (BD Biosciences, San Jose, CA, USA) at a densityof 5 × 105 cells/cm2 in serum-free DMEM with L-gluta-mine (Cellgro, Manassas, VA, USA) containing 50 μg/mL gentamicin (Sigma, St. Louis, MO, USA) at 37 C,5% CO2 to promote monocyte attachment. After 2hours, heat-inactivated fetal bovine serum (FBS, Invitro-gen, Carlsbad, CA, USA) was added to a final concentra-tion of 10%. Cells were >95% CD14+ as determined byFACS analysis (data not shown) prior to culture.
Macrophage stimulationAfter 5 days, the media and non-adhered cells wereremoved and replaced with complete media (DMEM,10% FBS, gentamicin) and incubated at 37 C, 5% CO2.Media was replaced every 2 days. Experiments wereinitiated upon confirmation of macrophage differentia-tion after 7 days in culture [12,20,21]. Macrophageswere used between day 7 and 10 and pretreated witheither: 1) 15 μg/mL of PHT-Na+ (Sigma), (serum levels,[22-24]), 2) 50 μg/mL PHT-Na+ (high dose), 3) 15 μg/mL PHT metabolite (Sigma), (5-(4’-hydroxyphenyl),5-phenylhydantoin, HPPH), or 4) 50 μg/mL HPPH for 1hour. Untreated cells served as control cultures. Stocksolutions of PHT-Na+ (150 mg/mL) were made in ster-ile deionized water while HPPH (150 mg/mL) solutionswere made in DMSO. Each stock solution was furtherdiluted prior to use. The total concentration of DMSOin cultures was always less than 0.05%. DMSO concen-trations less than 0.1% have been reported not to affectcellular viability and function [25,26]. Nevertheless, weconfirmed these findings in preliminary studies exposingmacrophage cultures to 0.05% DMSO (data not shown).To induce production of MMPs and proinflammatorycytokines, macrophages were challenged with 100 ng/mL purified LPS from the Gram-negative, periodontalpathogen, Aggregatibacter actinomycetemcomitans(A. actinomycetemcomitans (Aa), serotype b, strain Y4, akind gift from K. L. Kirkwood, University of South Car-olina, USA) for 24 hours. Isolation and purification ofAa LPS has been previously described [27]. Previous stu-dies have demonstrated that LPS from this organism iscapable of inducing MMP and TIMP production [28-30]and our preliminary studies determined that this con-centration of LPS was capable of significantly inducingTNF-a levels in human primary macrophages and THP-1 cells induced for macrophage differentiation (data notshown).
MMP, TIMP protein assaysAfter 24 hours, the media was collected, spun at 12,000 ×g, transferred to fresh tubes and stored at -80 C untilfurther use. Quantification of supernatant MMP and
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 2 of 10
TIMP levels were determined using the Luminex 100System (Luminex Co., Austin, TX, USA) and the Fluoro-kine MAP Multiplex Human MMP Panel and the Fluoro-kine MAP Human TIMP Multiplex Kit, respectivelyaccording to the manufacturer’s instructions (both fromR & D, Minneapolis, MN, USA). These kits measurelevels of pro-, mature, and TIMP complexed MMPs. Sixindependent experiments were performed from cellsderived from 6 different donors. The assays were per-formed in 96-well plates, as previously described [20].For MMP determination, microsphere beads coated withmonoclonal antibodies against MMP-1, MMP-2, MMP-3, MMP-9, MMP-12 were added to the wells. For TIMPdetermination, microsphere beads coated with monoclo-nal antibodies against TIMP-1, TIMP-2, TIMP-3, andTIMP-4 were added to the wells of a separate plate. Toremain below the upper level of quantitation, samplescontaining LPS were diluted 10-fold prior to analysis.This dilution factor was based on our preliminary studies.Samples and standards were pipetted into wells, incu-bated for 2 hours with the beads then washed using avacuum manifold (Millipore Corporation, Billerica, MAUSA). Biotinylated secondary antibodies were added andincubation for 1 h. The beads were then washed andincubated for an additional 30 minutes with streptavidinconjugated to the fluorescent protein, R-phycoerythrin(streptavidin/R-phycoerythrin). The beads were washedand analyzed (a minimum of 50 per analyte) using theLuminex 100 system. The Luminex 100 measures theamount of fluorescence associated with R-phycoerythrin,reported as median fluorescence intensity of each spec-tral-specific bead allowing it to distinguish the differentanalytes in each well. The concentrations of the unknownsamples (antigens in macrophage supernatants) were esti-mated from the standard curve using a third-order poly-nomial equation and expressed as pg/mL after adjustingfor the dilution factor. Samples below the detection limitof the assay were recorded as zero. The minimum detect-able concentrations for the assays were as follows: MMP-1: 4.4 pg/mL, MMP-2: 25.4 pg/mL, MMP-3: 1.3 pg/mL,MMP-9: 7.4 pg/mL, TIMP-1: 1.54 pg/mL, TIMP-2: 14.7pg/mL, TIMP-3: 86 pg/mL and TIMP-4: 1.29 pg/mL. Allvalues were standardized for total protein using the Brad-ford assay (Pierce, Thermo Scientific, Rockford, IL, USA)according to manufacturer’s instructions. Briefly, culturesupernatants were mixed with assay reagent and incu-bated for 10 minutes at room temperature in 96 wellplates. Bovine serum albumin (BSA, Invitrogen) was usedas a standard. The absorbance at 595 nm was read usinga SpectraMax M2 microplate reader (Molecular Devices,Sunnyvale, CA, USA). Values obtained from untreatedcontrol cultures were arbitrarily used as a baseline mea-sure. The ratio, (control)/(supernatant protein value) wasused to normalize each sample based on total protein.
Cytokine assaysAfter 24 hours, supernatants (n = 6 independentdonors) were collected and levels of TNF-a and IL-6determined by ELISA (RayBiotech, Norcross, GA, USA)according to manufacturer’s instructions. The absor-bance at 450 nm was read using a SpectraMax M2microplate reader (Molecular Devices) with the wave-length correction set at 550 nm. The rated sensitivitiesof the commercial ELISA kits was 15 pg/mL for TNF-aand 6 pg/mL for IL-6. Values were standardized fortotal protein using the Bradford assay as describedabove.
Cell viability assaysViability of macrophages was evaluated using the CellTi-ter 96 AQueous One Solution Cell Proliferation Assay[3-(4,5-diethylthiazol-2-yl)-5-(3-carboxymethoxyphenyl)-2-(4-sulfophenyl)-2H-tetrazolium, inner salt, MTS] assayaccording to the manufacturer’s protocol (Promega,Madison, WI, USA). This colorimetric method can beused to determine the number of viable cells in prolif-eration or to evaluate cytotoxicity. Briefly, macrophageswere cultured in triplicate in 96-well plates and treatedwith PHT, HPPH and LPS as described above. Unstimu-lated cells served as control cultures. After 24 h, thecells were incubated with MTS for 2 h at 37 C, 5% CO2.The absorbance was read at 490 nm using a microplatereader.
Statistical analysisData were analyzed using a hierarchical multiple regres-sion approach relative to LPS, drug and dose. The firsttier sought to establish the validity of the positive con-trol, LPS vs the negative control group. The second tierof this analysis was aimed at determining whether PHTor HPPH have an effect on MMP, TIMP, TNF-a andIL-6 levels. Finally, the third tier sought to contrast doseand compare one drug with another. Data wereexpressed as mean ± SEM and compared using a two-tailed Student’s t test for correlated samples (GraphPadPrism, GraphPad Software, La Jolla, CA, USA). Resultswere considered statistically significant at p < 0.05.
ResultsPHT and HPPH inhibit LPS-induced supernatant levels ofMMP-1, MMP-3, MMP-9, and TIMP-1 in a dose dependentmannerTo evaluate the effects of PHT and its metabolite,HPPH on macrophage MMP and TIMP levels, humanmonocyte-derived macrophages were pretreated for 1hour with either 15 μg/mL or 50 μg/mL of these agentsprior to challenge with LPS. Previous studies have deter-mined that PHT plasma levels of 10-20 μg/mL arenecessary to effectively maintain effective seizure control
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 3 of 10
[22-24]. Thus, the concentrations used in our studyrepresent therapeutic as well as elevated levels of PHTpermitting the evaluation of dose on MMP and TIMPproduction. To rule out the possibility that differencesin supernatant levels of these readouts were due todecreased cell viability, we performed a viability assayon cells cultured in each condition. No significant differ-ences were noted in the viability of cells exposed to LPSand either dose of PHT, HPPH, PHT/LPS or HPPH/LPS as determined by MTS assay. Further, we standar-dized the results of each analyte to total protein concen-tration for each condition using a Bradford assay. Nodifferences were noted for any analyte examined in con-ditioned media from macrophage cultures treated with
PHT or HPPH alone compared to control cultures (p >0.05). As expected, LPS markedly induced supernatantMMP-1, MMP-3, MMP-9, TIMP-1 but not MMP-2levels in our 6 independent cultures after a 24 hourexposure (Fig. 1A-D). Compared to untreated controlcultures, LPS significantly increased secretion of MMP-1despite the presence of either PHT or HPPH at anydose. This was similarly observed for MMP-3 levels withthe exception of cultures pretreated with 15 μg/mLHPPH which despite elevated levels, did not reach sta-tistical significance (p > 0.05). In contrast, exposure ofmacrophages to 50 μg/mL of either PHT or HPPH priorto LPS stimulation prevented a significant increase inMMP-9 and TIMP-1 (Fig. 1D and Fig. 2). Levels of
Figure 1 The effect of phenytoin, HPPH and LPS on levels of (A) matrix metalloproteinase-1, (B) matrix metalloproteinase-2, (C) matrixmetalloproteinase-3, and (D) matrix metalloproteinase-9 in conditioned medium from macrophage cultures. Primary human monocyte-derived macrophages (n=6 independent cultures) were pretreated with phenytoin or HPPH (15 μg/mL and 50 μg/mL) for 1 hour prior tochallenge with 100 ng/mL A. actinomycetemcomitans LPS and the levels of matrix metalloproteinase-1, matrix metalloproteinase-2, matrixmetalloproteinase-3, and matrix metalloproteinase-9 measured after 24 hours in conditioned media by multiplex analysis. MMP-1, matrixmetalloproteinase-1; MMP-2, matrix metalloproteinase-2; MMP-3, matrix metalloproteinase-3; MMP-9, matrix metalloproteinase-9; CON, control;PHT, phenytoin; HPPH, 5-(4-hydroxyphenyl-),5- phenylhydantoin; LPS, lipopolysaccharide. Compared to CON, # p<0.05, ## p<0.01, ### p<0.001,compared to LPS, * p<0.05, ** p<0.01, *** p<0.001. Student t-test, n=6 independent donors.
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 4 of 10
MMP-9 and TIMP-1 remained near control levelsdespite the potent proinflammatory challenge thusdemonstrating the ability of these agents to alter macro-phage function. Compared to LPS alone, pretreatmentwith 50 μg/mL PHT significantly blunted LPS-inducedlevels of MMP-1 (p < 0.05). In cultures pretreated with50 μg/mL HPPH, MMP-3 levels were not significantlydifferent compared to LPS-only treated cultures (p >0.05) although the trend for reduced supernatant levelsof MMP3 was evident. However, exposure of macro-phages to either 15 μg/mL or 50 μg/mL PHT prior toLPS stimulation significantly blunted supernatant MMP-3 levels (p < 0.01 and p < 0.001, respectively, Fig. 1C)compared to LPS-only treated cultures. Interestingly, atrend for higher levels of MMP-1 were noted in culturestreated with HPPH while MMP-3 levels were slightlyelevated in cultures treated with either PHT and HPPHalthough neither reached statistical significance (p >0.05) (Fig. 1, A, C).Elevated levels (50 μg/mL) of PHT or HPPH signifi-
cantly reduced MMP-9 and TIMP-1 levels compared toLPS-only treated cells (Fig. 1D and Fig. 2). The levels ofthese analytes remained near control values despite LPSchallenge. Interestingly, HPPH but not PHT was
associated with reduced levels of MMP-2 compared toLPS only, but this relationship was not statistically sig-nificant. MMP-12 and TIMPs-2-4 remained below levelsof detection in all groups and cultures.
Supernatant levels of TNF-a but not IL-6, is decreased inresponse to PHTAt 24 hours, supernatant levels of TNF-a and IL-6were significantly increased by LPS compared tountreated controls (p < 0.001). Similar to MMP andTIMP levels, no significant differences in TNF-a andIL-6 levels were observed in supernatants exposed toeither 15 or 50 μg/mL PHT and HPPH alone com-pared to untreated cultures although a trend fordecreased levels of TNF-a was evident (Fig 3A). How-ever, macrophage cultures pretreated with 50 μg/mLPHT prior to challenge with LPS showed a significant(p < 0.05) decrease in TNF-a levels compared to LPSonly treated cultures. No difference was noted for 15μg/mL of PHT or HPPH at either concentration (Fig.3A). Regardless of dosage, pretreatment with PHT orHPPH prior to LPS challenge had no significant effect(p > 0.05) on IL-6 secretion when compared to LPSonly treated cultures.
Figure 2 The effect of phenytoin, HPPH and LPS on levels of tissue inhibitor of matrix metalloproteinase-1 in conditioned mediumfrom macrophage cultures. Primary human monocyte-derived macrophages (n = 6 independent cultures) were pretreated with phenytoin orHPPH (15 μg/mL and 50 μg/mL) for 1 hour prior to challenge with 100 ng/mL A. actinomycetemcomitans LPS and the levels of tissue inhibitor ofmatrix metalloproteinase-1 measured after 24 hours in conditioned media by multiplex analysis. TIMP-1, tissue inhibitor of matrixmetalloproteinase-1; CON, control; PHT, phenytoin; HPPH, 5-(4-hydroxyphenyl-),5-phenylhydantoin; LPS, lipopolysaccharide. Compared to CON, #p < 0.05, ## p < 0.01, ### p < 0.001, compared to LPS, * p < 0.05, ** p < 0.01, *** p < 0.001. Student t-test, n = 6 independent donors.
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 5 of 10
DiscussionMacrophages are involved in a remarkably diverse arrayof homeostatic processes of vital importance to the host.In addition to their critical role in immunity [12],macrophages are also widely recognized as ubiquitousmediators of cellular turnover and maintenance of extra-cellular matrix homeostasis [13-18]. However, beyondtheir essentiality in immunity and tissue homeostasis,the macrophage has also been implicated in the evolu-tion of periodontal pathological processes including per-iodontal disease and DIGO [11,19,20,31,32]. Thisinvestigation posited that macrophage-derived expres-sion of proinflammatory cytokines, MMPs and/or TIMPexpression is blunted upon exposure to PHT and/orHPPH hindering the ability of these cells to contributeto the fibroblast-mediated degradation of exuberantECM proteins seen in DIGO. Since plaque-induced gin-gival inflammation exacerbates the manifestations ofPHT-induced GO [33], we exposed macrophage culturesto purified LPS from the periodontal pathogen, A. acti-nomycetemcomitans (Aa) and examined protein levels ofMMPs, TIMPs and proinflammatory cytokines in condi-tioned media. Aa can be isolated from plaque samplesof patients with GO [34] while Aa LPS, a TLR4 agonist,strongly induces MMP and pro-inflammatory cytokineexpression [28-30,35].We exposed macrophage cultures to 2 different con-
centrations of PHT and HPPH. And while PHT plasmalevels of 10-20 μg/mL are necessary to effectively main-tain effective seizure control [22-24], disturbances in
plasma as well as gingival concentrations of PHT arelikely associated with DIGO. Indeed, Güncü et al [36]compared PHT levels in plasma and gingival crevicularfluid (GCF), a serum exudate, from subjects whodemonstrated gingival overgrowth (responders) vs. thosewho did not (non-responders). Although PHT wasdetected in all of the GCF and plasma samples, themean concentration of PHT was significantly greater inGCF compared to plasma (294.99 ± 430.15 μg/mL vs.16.09 ± 4.21 μg/mL, respectively). Further, the concen-tration of plasma PHT was significantly higher inresponders compared to non-responders (16.09 ± 4.21μg/mL vs. 9.93 ± 4.56, respectively).MMP-1 is recognized as an important mediator of
connective tissue remodeling reported to be present athigh concentrations in inflamed gingiva [37]. In the pre-sent study, supernatant MMP levels did not demonstrateany significant differences in response to PHT andHPPH alone at either dose compared to untreatedmacrophage cultures although we noted a trend forhigher levels of MMP-1 and MMP-3. This finding wasattributed to donor-specific variations in responses tothese agents and serve to highlight clinical observationsthat approximately 50% of patients taking PHT developGO [1,2]. This notion is supported by the finding thatfibroblasts derived from subjects with cyclosporine-A(CSA)-induced gingival overgrowth produce significantlylower levels of MMP-1 than fibroblasts derived fromsubjects without overgrowth [38]. In the present study,supernatant levels of several MMPs were significantly
Figure 3 The effect of phenytoin, HPPH and LPS on levels of (A) TNF-a and (B) IL-6 in conditioned medium from macrophagecultures. Primary human monocyte-derived macrophages (n = 6 independent cultures) were pretreated with phenytoin or HPPH (15 μg/mLand 50 μg/mL) for 1 hour prior to challenge with 100 ng/mL A. actinomycetemcomitans LPS and the levels of TNF-a and IL-6 measured inconditioned media after 24 hours by enzyme-linked immunosorbent assay (ELISA). TNF-a, tumor necrosis-alpha; IL, interleukin; CON, control; PHT,phenytoin; HPPH, 5-(4-hydroxyphenyl-), 5-phenylhydantoin; LPS, lipopolysaccharide. Compared to CON, # p < 0.05, ## p < 0.01, ### p < 0.001,compared to LPS, * p < 0.05, ** p < 0.01, *** p < 0.001. Student t-test, n = 6 independent donors.
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 6 of 10
decreased relative to LPS-only cultures in a dose-depen-dent manner suggesting that PHT and HPPH may miti-gate the macrophage’s ability to degrade ECM proteinsby limiting its natural response to produce metallopro-teinases. Such a dose response is consistent with otherstudies which have demonstrated, not only a similareffect on MMP-1 and MMP-3 at the protein andmRNA level [39-43], but also that a threshold of serumconcentration of CSA helps to govern this mechanism[44-49].MMP activity is counteracted by the actions of
TIMPs. Here we report that exposure of macrophagesto LPS was associated with an increase in TIMP-1 levelswhile exposure to high concentration (50 μg/mL) ofPHT and HPPH, on the other hand, significantlyreduced TIMP-1 levels. This finding is in agreementwith in-vitro and in-vivo studies which report a relativereduction in MMP-1 and MMP-8/TIMP-1 in gingivalfibroblasts and in serum and GCF concentration inCSA-associated gingival overgrowth subjects [50,51].This reflects more a decrease in MMP production ratherthan an increase in TIMP. In fact, this corresponds withour findings in that supernatant levels of TIMP-1 insamples treated with both LPS and high doses of PHTor HPPH were not significantly different relative tountreated controls (Fig. 2). The net effect on ECMmetabolism is based on the relative ratios of MMP andTIMP. When MMP levels decrease and/or TIMP levelsincrease, the turnover of ECM diminishes, potentiallyleading to an exuberant accumulation of these proteins.In this study, elevated levels of PHT in LPS-stimulatedmacrophages were associated with decreases in bothMMP and TIMP levels. Therefore the decrease inTIMP-1 levels was counteracted by decreases in MMPlevels. As a result, the macrophage’s synergistic relation-ship with the fibroblast would be compromised leadingto DIGO. Indeed, monocytes (macrophage precursors)can stimulate fibroblasts to produce MMP-1 by cell-cellinteractions while conditioned media from monocytes iscapable of inducing MMP-1 production in fibroblasts[52]. How PHT and HPPH impact monocyte/macro-phage-fibroblast interactions and MMP productionrequires further study.PHT is known to affect Na+ as well as Ca2+ metabo-
lism [3], (e.g., Ca2+ channels) and it is likely that thiswill impact MMP/TIMP and cytokine levels [53].Indeed, Na+ channels have been linked to activation ofmacrophages and microglia [54] and accumulating evi-dence indicates that sodium channel blockers can con-tribute to modulation of immune functions [55]. PHThas been reported to ameliorate the inflammatoryresponse associated with experimental autoimmuneencephalomyelitis in mice [54], modulate intracellularsignaling cascades to TLR ligands [56] and significantly
reduce LPS-induced phagocytosis in-vitro[53]. Here wereport a dose-dependent inhibition of macrophage func-tion by way of suppressed supernatant levels of MMP-1,MMP-3, MMP-9, TIMP-1 and TNF-a by PHT inhuman macrophages challenged with LPS. PHT hasbeen reported to inhibit both activation of T-type cal-cium channels and RANKL-induced expression of c-fosprotein in bone marrow-derived macrophages implyingthat calcium signals play a role in c-fos expression [57].PHT was also shown to inhibit NFATc1 signaling inthese cells. Further, in atrial myocytes, pharmacologicalinhibition of NFAT with 11R-VIVIT almost completelyblunted the stretch-induced up-regulation of active-MMP-2/-9 [58]. Kiode et al [57] suggested that PHTmay inhibit NFATc1 signals through suppression of c-fos expression. Since c-fos/AP-1 regulates the expressionof numerous inflammatory cytokines and MMPs/TIMPsvia promoter AP-1 binding motif [59,60], suppression ofc-fos may provide a possible mechanism wherebyMMPs/TIMPs and possibly cytokine levels are inhibited.In contrast to PHT we report a dose-dependent inhi-
bition of MMP-9 and TIMP-1 by HPPH in cultureschallenged with LPS. These discrepancies may be attrib-uted to differences in the interactions of these drugswith target molecules. Kobayashi et al [61] reported thatPHT and 5-(4-methylphenyl)-5-phenylhydantoin, whichcontain a phenyl or methylphenyl group at both R2-and R3-positions activated the ligand binding domain ofhuman pregane X receptor (hPXR), a member of thenuclear receptor family of ligand-activated transcrip-tional factors, whereas 5-(4-hydroxyphenyl)-5-phenylhy-dantoin did not. Alternatively, it is possible that higherconcentrations of HPPH may be required to achieveresults similar to that observed with PHT as evident bythe trend for blunting of MMP-1 and MMP3 at higherdoses of HPPH (Fig. 1A, C). Nevertheless, these findingsserve to highlight the impact of PHT and HPPH, on themacrophage’s ability to contribute to ECM turnover andunderscore the importance of Na+ and Ca2+ channels inactivated macrophages.An interesting finding of our study was the suppres-
sion of TNF-a but not IL-6 by PHT. IL-6 enhances pro-liferation of fibroblasts and exerts a positive effect oncollagen and glycosaminoglycan synthesis [62,63]. Athigh levels, TNF-a has been reported to inhibit collagensynthesis [64] and increase MMP synthesis in gingivalfibroblasts [65-67], which contributes to gingival break-down. Conversely, at low levels (< 10 ng/ml) TNF-a sti-mulates cellular proliferation, induces production ofECM and inhibits phagocytosis of collagen by gingivalfibroblasts [68,69]. Since TNF-a enhances MMP-1 [70]and MMP-9 [71] expression, the blunting of TNF-alevels observed in the present study may have contribu-ted to the decrease in supernatant levels of MMP-1 and
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
Page 7 of 10
MMP-9. In microglial cells, blockade of sodium chan-nels with PHT significantly reduced LPS-induced secre-tion of IL-1a, IL-1b, and TNF-a, but not IL-6 or IL-10suggesting that sodium channels participate in the pro-cess of cytokine release [53]. In agreement, we notedspecific modulation of LPS-induced TNF-a but not IL-6in the presence of high concentrations of PHT (50 μg/ml). Black et al [53] also demonstrated that tetrodo-toxin, a sodium channel blocker, inhibited secretion ofIL-1a, IL-1b, and TNF-a secretion but to a lesserdegree than PHT, in spite of similar inhibitory actionson sodium channels. This difference was likely due tothe effects on Ca2+ metabolism by PHT. It was alsointeresting to note that HPPH had no effect on TNF-alevels. As discussed above this may be due to differencesin the interactions of HPPH with target molecules orthat higher dose of HPPH is required for inhibition ofTNF-a.In macrophages, increased TNF-a production in
response to LPS challenge is associated with a transi-ent increase in intracellular calcium [72,73] so thatintracellular calcium may participate as a second mes-senger in TLR4-dependent signaling [72,74]. Insightinto a possible mechanism linking intracellular calciumand cytokine levels was recently demonstrated usingRAW macrophages [75]. Using a pharmacologicalapproach, Yamashiro et al [75] examined the role oftransient receptor potential vanilloin 4 (TRPV2), a cal-cium permeable channel, in LPS-induced calciummobilization and induction of cytokines. They reportedthat LPS-induced IL-6 production was due at least inpart by calcium mobilization solely from intracellularsources and partly by entry of extracellular calciumthrough TRPV2. Further, they reported that in additionto calcium mobilization through the IP3-receptor,TRPV2-mediated intracellular calcium mobilizationinvolved NF�B-dependent TNF-a and IL-6 expression,while extracellular calcium entry is involved in NF�B-independent IL-6 production. Collectively, these find-ings may provide insights into how PHT and HPPHmodulate cytokine and possibly MMP/TIMP levels.Future studies will be necessary to evaluate the impactof these agents on intracellular and extracellular cal-cium levels in macrophages prior to LPS challenge andtheir correlation to cytokine and MMP/TIMPproduction.
ConclusionsOur results demonstrate that PHT as well as its metabo-lite, HPPH significantly blunt A. actinomycetemcomitansLPS-induced levels of MMP-1, MMP-3, MMP-9 andTIMP-1 in a dose-dependent manner and that a highconcentration of PHT significantly decreases TNF-a butnot IL-6 levels in the human macrophage. Given the
presence of significant numbers of macrophages in gin-gival tissues and the correlation between the quality ofplaque control and fibrosis, our data reveals a mechan-ism whereby both PHT and its metabolite, HPPH dysre-gulate macrophage function. Blunting of macrophagederived MMPs and TNF-a by these agents in responseto stimuli may permit collagen accumulation withoutthe counteracting production of MMPs by these cells.
AcknowledgementsWe would like to thank Dr. Keith L. Kirkwood (University of South Carolina,USA) for his kind gift of purified A. actinomycetemcomitans LPS and to Dr.Steven Offenbacher and Dr. Silvana Barros (University of North Carolina atChapel Hill) for helpful suggestions to this report. We would also like tothank Janice Ko and Roger Arce for their technical assistance. This work wassupported by the University of North Carolina at Chapel Hill, School ofDentistry, North Carolina, USA.
Author details1Department of Periodontology, School of Dentistry, University of NorthCarolina at Chapel Hill, Chapel Hill, North Carolina, USA. 2Department ofPeriodontics, School of Dentistry, Loma Linda University, Loma Linda, CA,92350, USA.
Authors’ contributionsRS, NA and SN contributed to the concept and design of the study, and tothe manuscript writing. SN, RS and AA performed isolated of monocytes andculture of macrophages. RS and AA performed the MMP, TIMP proteinassays, cytokine assays, and viability assays. RS, NA and SN performed thedata analysis. All authors read and approved the final manuscript.
Competing interestsThe authors declare that they have no competing interests.
Received: 30 April 2010 Accepted: 15 September 2010Published: 15 September 2010
3. Pincus HH: Diphenyldantoin and ion flux in lobster nerve. Arch Neur 1972,26:4-10.
4. Bajpai M, Roskos LK, Shen DD, Levy RH: Roles of cytochrome P4502C9 andcytochrome P4502C19 in the stereoselective metabolism of phenytointo its major metabolite. Drug Metab Dispos 1996, 24:1401-1403.
5. Lin CJ, Yen MF, Hu OY, Lin MS, Hsiong CH, Liou HH: Association ofgalactose single-point test levels and phenytoin metabolicpolymorphisms with gingival hyperplasia in patients receiving long-termphenytoin therapy. Pharmacotherapy 2008, 28:35-41.
6. Ieiri I, Goto W, Hirata K, Toshitani A, Imayama S, Ohyama Y, Yamada H,Ohtsubo K, Higuchi S: Effect of 5-(p-hydroxyphenyl)-5-phenylhydantoin(p-HPPH) enantiomers, major metabolites of phenytoin, on theoccurrence of chronic-gingival hyperplasia: in vivo and in vitro study.Eur J Clin Pharmacol 1995, 49:51-56.
11. Iacopino A, Doxey D, Cutler C, Nares S, Stoever K, Fojt J, Gonzales A, Dill RE:Phenytoin and cyclosporine A specifically regulate macrophagephenytoin and expression of platelet-derived growth factor andinterleukin-1 in vitro and in vivo: possible molecular mechanism ofdrug-induced gingival hyperplasia. J Periodontol 1997, 68:73-83.
12. Nares S, Wahl SM: Monocytes and Macrophages. In Measuring Immunity:Basic Science and Clinical Practice. Edited by: Lotze MT, Thomson AT.London: Elsevier; , 1 2005:299-311.
13. Riches DW: The multiple roles of macrophages in wound healing. In TheMolecular and Cellular Biology of Wound Repair. Edited by: Clark RAF,Henson PM. New York: Plenum Press; 1988:213-239.
25. Chen RM, Chen TG, Chen TL, Lin LL, Chang CC, Chang HC, Wu CH: Anti-inflammatory and antioxidative effects of propofol onlipopolysaccharide-activated macrophages. Ann N Y Acad Sci 2005,1042:262-271.
26. Rival Y, Benéteau N, Chapuis V, Taillandier T, Lestienne F, Dupont-Passelaigue E, Patoiseau JF, Colpaert FC, Junquéro D: Cardiovascular drugsinhibit MMP-9 activity from human THP-1 macrophages. DNA Cell Biol2004, 23:283-292.
27. Rossa C Jr, Liu M, Bronson P, Kirkwood KL: Transcriptional activation ofMMP-13 by periodontal pathogenic LPS requires p38 MAP kinase. JEndotoxin Res 2007, 13:85-93.
28. Bodet C, Chandad F, Grenier D: Inflammatory responses of amacrophage/epithelial cell co-culture model to mono and mixedinfections with Porphyromonas gingivalis, Treponema denticola, andTannerella forsythia. Microbes Infec 2006, 8:27-35.
29. Woo CH, Lim JH, Kim JH: Lipopolysaccharide induces matrixmetalloproteinase-9 expression via a mitochondrial reactive oxygenspecies-p38 kinase-activator protein-1 pathway in Raw 264.7 cells. JImmunol 2004, 173:6973-6980.
30. Zhang Y, Mc Cluskey K, Fujii K, Wahl L: Differential regulation of monocytematrix metalloproteinase and TIMP-1 production by TNF-alpha,granulocyte-macrophage CSF, and IL-1 beta through prostaglandin-dependent and -independent mechanisms. J Immunol 1998,161:3071-3076.
31. Trackman PC, Kantarci A: Connective tissue metabolism and gingivalovergrowth. Crit Rev Oral Biol Med 2004, 4:165-175.
34. Akiyama S, Amano A, Kato T, Takada Y, Kimura KR, Morisaki I: Relationshipof periodontal bacteria and Porphyromonas gingivalis fimA variationswith phenytoin-induced gingival overgrowth. Oral Dis 2006, 12:51-56.
35. La VD, Bergeron C, Gafner S, Grenier D: Grape seed extract suppresseslipopolysaccharide-induced matrix metalloproteinase (MMP) secretion bymacrophages and inhibits human MMP-1 and -9 activities. J Periodontol2009, 80:1875-1882.
36. Güncü GN, Caglayan F, Dinçel A, Bozkurt A, Saygi S, Karabulut E: Plasmaand gingival crevicular fluid phenytoin concentrations as risk factors forgingival overgrowth. J Periodontol 2006, 77:2005-2010.
37. Ryan ME, Golub LM: Modulation of matrix metalloproteinase activities inperiodontitis as a treatment of strategy. J Periodontol 2000, 24:226-238.
38. Sukkar TZ, Thomason JM, Cawston TE, Lakey R, Jones D, Catterall J,Seymour RA: Gingival fibroblasts grown from cyclosporin-treatedpatients show a reduced production of matrix metalloproteinase-1(MMP-1) compared with normal gingival fibroblasts, and cyclosporindown-regulates the production of MMP-1 stimulated by pro-inflammatory cytokines. J Periodontal Res 2007, 42:580-588.
39. Sugano N, Ito K, Murai S: Cyclosporin A inhibits collagenase geneexpression via AP-1 and JNK suppression in human gingival fibroblasts.J Periodontal Res 1998, 33:448-452.
40. Bolzani G, Della Coletta R, Martelli Júnior H, Martelli Júnior H, Graner E:Cyclosporin A inhibits production and activity of matrixmetalloproteinases by gingival fibroblasts. J Periodontal Res 2000,35:51-58.
41. Kataoka M, Shimizu Y, Kunikiyo K, Asahara Y, Yamashita K, Ninomiya M,Morisaki I, Ohsaki Y, Kido JI, Nagata T: Cyclosporin A decreases thedegradation of type I collagen in rat gingival overgrowth. J Cell Physiol2000, 182:351-358.
42. Yamada H, Nishimura F, Naruishi K, Chou HH, Takashiba S, Albright GM,Nares S, Iacopino AM, Murayama Y: Phenytoin and cyclosporin A suppressthe expression of MMP-1, TIMP-1, and cathepsin L, but not cathepsin Bin cultured gingival fibroblasts. J Periodontol 2000, 71:955-960.
43. Hyland PL, Traynor PS, Myrillas TT, Marley JJ, Linden GJ, Winter P,Leadbetter N, Cawston TE, Irwin CR: The effects of cyclosporin on thecollagenolytic activity of gingival fibroblasts. J Periodontol 2003,74:437-445.
44. McGaw T, Lam S, Coates J: Cyclosporin-induced gingival overgrowth:correlation with dental plaque scores, gingivitis scores, and cyclosporinlevels in serum and saliva. Oral Surg Oral Med Oral Pathol 1987,64:293-297.
45. Pan WL, Chan CP, Huang CC, Lai MK: Cyclosporine-induced gingivalovergrowth. Transplant Proc 1992, 24:1393-1394.
46. Pernu HE, Pernu LM, Huttunen KR, Nieminen PA, Knuuttila ML: Gingivalovergrowth among renal transplant recipients related toimmunosuppressive medication and possible local background factors. JPeriodontol 1992, 63:548-553.
47. King GN, Fullinfaw R, Higgins TJ, Walker RG, Francis DM, Wiesenfeld D:Gingival hyperplasia in renal allograft recipients receiving cyclosporin-Aand calcium antagonists. J Clin Periodontol 1993, 20:286-293.
48. O’Valle F, Mesa FL, Gómez-Morales M, Aguilar D, Caracuel MD, Medina-Cano MT, Andújar M, López-Hidalgo J, García del Moral R:Immunohistochemical study of 30 cases of cyclosporin A-inducedgingival overgrowth. J Periodontol 1994, 65:724-730.
49. Thomason JM, Seymour RA, Ellis JS, Kelly PJ, Parry G, Dark J, Idle JR:Iatrogenic gingival overgrowth in cardiac transplantation. J Periodontol1995, 66:742-746.
50. Emingil G, Afacan B, Tervahartiala T, Toz H, Atilla G, Sorsa T: Gingivalcrevicular fluid and serum matrix metalloproteinase-8 and tissueinhibitor of matrix metalloproteinase-1 levels in renal transplant patientsundergoing different immunosuppressive therapy. J Clin Periodontol 2008,35:221-229.
51. Gagliano N, Moscheni C, Dellavia C, Stabellini G, Ferrario VF, Gioia M:Immunosuppression and gingival overgrowth: gene and proteinexpression profiles of collagen turnover in FK506-treated humangingival fibroblasts. J Clin Periodontol 2005, 32:167-173.
52. Domeij H, Yucel-Lindberg T, Modéer T: Cell interactions between humangingival fibroblasts and monocytes stimulate the production of matrixmetalloproteinase-1 in gingival fibroblasts. J Periodontal Res 2006,41:108-117.
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
53. Black JA, Liu S, Waxman SG: Sodium channel activity modulates multiplefunctions in microglia. Glia 2009, 57:1072-1081.
54. Craner MJ, Damarjian TG, Liu S, Hains BC, Lo AC, Black JA, Newcombe J,Cuzner ML, Waxman SG: Sodium channels contribute to microglia/macrophage activation and function in EAE and MS. Glia 2005,49:220-229.
56. Suzuki AM, Yoshimura A, Ozaki Y, Kaneko T, Hara Y: Cyclosporin A andphenytoin modulate inflammatory responses. J Dent Res 2009,88:1131-1136.
57. Koide M, Kinugawa S, Ninomiya T, Mizoguchi T, Yamashita T, Maeda K,Yasuda H, Kobayashi Y, Nakamura H, Takahashi T, Udagawa N:Diphenylhydantoin inhibits osteoclast differentiation and functionthrough suppression of NFATc1 signaling. J Bone Miner Res 2009,24:1469-1480.
58. Saygili E, Rana OR, Meyer C, Gemein C, Andrzejewski MG, Ludwig A,Weber C, Schotten U, Krüttgen A, Weis J, Schwinger RH, Mischke K, Rassaf T,Kelm M, Schauerte P: The angiotensin-calcineurin-NFAT pathwaymediates stretch-induced up-regulation of matrix metalloproteinases-2/-9 in atrial myocytes. Basic Res Cardiol 2009, 104:435-448.
60. Clark IM, Swingler TE, Sampieri CL, Edwards DR: The regulation of matrixmetalloproteinases and their inhibitors. Int J Biochem Cell Biol 2008,40:1362-1378.
61. Kobayashi K, Yamagami S, Higuchi T, Hosokawa M, Chiba K: Key structuralfeatures of ligands for activation of human pregnane X receptor. DrugMetab Dispos 2004, 32:468-472.
62. Ramsden L, Rider CC: Selective and differential binding of interleukin(1L)-1 alpha, IL-1 beta, IL-2, and IL-6 to glycosaminoglycans. Eur JImmunol 1992, 22:3027-3031.
64. Modeer T, Brunius G, Iinuma M, Lerner UH: Phenytoin potentiatesinterleukin-1 induced prostaglandin biosynthesis in human gingivalfibroblasts. Bri J Pharma 1992, 106:574-578.
65. Saren P, Welgus HG, Kovanen PT: TNF-alpha and IL-1b selectively induceexpression of 92-kDa gelatinase by human macrophages. J Immunol1996, 157:4159-4165.
66. Domeij H, Yucel-Lindberg T, Modeer T: Signal pathways involved in theproduction of MMP-1 and MMP-3 in human gingival fibroblasts. Eur JOral Sci 2002, 110:302-306.
67. Birkedal-Hansen H, Moore WG, Bodden MK, Birkedal-Hansen H: Matrixmetalloproteinease: a review. Crit Rev Oral Biol Med 1993, 4:197-250.
68. Sugarman BJ, Aggarwal BB, Hass PE, Figari IS, Palladino MA Jr, Shepard HM:Recombinant human tumor necrosis factor-alpha: effects onproliferation of normal and transformed cells in vitro. Science 1985,230:943-945.
69. Chou DH, Lee W, McCulloch CA: TNF-alpha inactivation of collagenreceptors: implications for fibroblast function and fibrosis. J Immunol1996, 156:4354-4362.
70. Reunanen N, Li S-P, Ahonen M, Foschi M, Han J, Kahari V-M: Activation ofp38alpha MAPK enhances collagenase-1 (matrix metalloproteinase(MMP)-1) and stromelysin-1 (MMP-3) expression by mRNA stabilization. JBiol Chem 2002, 277:32360-32368.
71. Srivastava AK, Qin X, Wedhas N, Arnush M, Linkhart TA, Chadwick RB,Kumar A: Tumor necrosis factor-{alpha} augments matrixmetalloproteinase-9 production in skeletal muscle cells through theactivation of transforming growth factor–activated kinase 1 (TAK1)-dependent signaling pathway. J Biol Chem 2007, 282:35113-35124.
72. Lichtman SN, Wang J, Zhang C, Lemasters JJ: Endocytosis and Ca2+ arerequired for endotoxin-stimulated TNF-alpha release by rat Kupffer cells.Am J Physiol 1996, 271:G920-G928.
73. Letari O, Nicosia S, Chiavaroli C, Vacher P, Schlegel W: Activation bybacterial lipopolysaccharide causes changes in the cytosolic free calciumconcentration in single peritoneal macrophages. J Immunol 1991,147:980-983.
74. Zhou X, Yang W, Li J: Ca2+- and protein kinase C-dependent signalingpathway for nuclear factor-kappa B activation, inducible nitric oxidesynthase expression, and tumor necrosis factor-alpha production in
lipopolysaccharide-stimulated rat peritoneal macrophages. J Biol Chem2006, 281:31337-31347.
75. Yamashiro K, Sasano T, Tojo K, Namekata I, Kurokawa J, Sawada N,Suganami T, Kamei Y, Tanaka H, Tajima N, Utsunomiya K, Ogawa Y,Furukawa T: Role of transient receptor potential vanilloid 2 in LPS-induced cytokine production in macrophages. Biochem Biophys ResCommun 2010, 398:284-289.
doi:10.1186/1476-9255-7-48Cite this article as: Serra et al.: Suppression of LPS-induced matrix-metalloproteinase responses in macrophages exposed to phenytoinand its metabolite, 5-(p-hydroxyphenyl-), 5-phenylhydantoin. Journal ofInflammation 2010 7:48.
Submit your next manuscript to BioMed Centraland take full advantage of:
• Convenient online submission
• Thorough peer review
• No space constraints or color figure charges
• Immediate publication on acceptance
• Inclusion in PubMed, CAS, Scopus and Google Scholar
• Research which is freely available for redistribution
Submit your manuscript at www.biomedcentral.com/submit
Serra et al. Journal of Inflammation 2010, 7:48http://www.journal-inflammation.com/content/7/1/48
![2-[( E )-(3-Carboxy-4-hydroxyphenyl)iminiomethyl]-4-chlorophenolate](https://static.documents.pub/doc/80x56/635b6e3026d04ababa024cc9/2-e-3-carboxy-4-hydroxyphenyliminiomethyl-4-chlorophenolate.jpg)
![FT-IR, FT-Raman spectroscopy and computational study of N-carbamimidoyl-4-{[(E)-((2-hydroxyphenyl)methylidene]amino}benzenesulfonamide](https://static.documents.pub/doc/80x56/6319c396bc8291e22e0f4222/ft-ir-ft-raman-spectroscopy-and-computational-study-of-n-carbamimidoyl-4-e-2-hydroxyphenylmethylideneaminobenzenesulfonamide.jpg)
![P3-[2-(4-hydroxyphenyl)-2-oxo]ethyl ATP for the Rapid Activation of the Na+,K+-ATPase](https://static.documents.pub/doc/80x56/6344695b596bdb97a90872c6/p3-2-4-hydroxyphenyl-2-oxoethyl-atp-for-the-rapid-activation-of-the-nak-atpase.jpg)
![( E )-4-Hydroxy-2-[(2-hydroxyphenyl)iminiomethyl]phenolate](https://static.documents.pub/doc/80x56/635b880026d04ababa0259b9/-e-4-hydroxy-2-2-hydroxyphenyliminiomethylphenolate.jpg)
![Bis{2-[( E )-(5- tert -butyl-2-hydroxyphenyl)diazenyl]benzoato}dimethyltin(IV)](https://static.documents.pub/doc/80x56/634790ce494dcf875e047cce/bis2-e-5-tert-butyl-2-hydroxyphenyldiazenylbenzoatodimethyltiniv.jpg)
